Update app.py

app.py

@@ -20,7 +20,7 @@ model_name = 'cognitivecomputations/dolphin-vision-72b'
 model = AutoModelForCausalLM.from_pretrained(
     model_name,
     torch_dtype=torch.float16,
-    device_map="auto",  # This will automatically use the GPU if available
+    device_map="auto",
     trust_remote_code=True
 )
 
@@ -29,7 +29,7 @@ tokenizer = AutoTokenizer.from_pretrained(
     trust_remote_code=True
 )
 
-def inference(prompt, image):
+def inference(prompt, image, temperature, beam_size):
     messages = [
         {"role": "user", "content": f'<image>\n{prompt}'}
     ]
@@ -55,6 +55,8 @@ def inference(prompt, image):
             input_ids,
             images=image_tensor,
             max_new_tokens=1024,
+            temperature=temperature,
+            num_beams=beam_size,
             use_cache=True
         )[0]
 
@@ -65,10 +67,16 @@ with gr.Blocks() as demo:
         with gr.Column():
             prompt_input = gr.Textbox(label="Prompt", placeholder="Describe this image in detail")
             image_input = gr.Image(label="Image", type="pil")
+            temperature_input = gr.Slider(minimum=0.1, maximum=2.0, value=0.7, step=0.1, label="Temperature")
+            beam_size_input = gr.Slider(minimum=1, maximum=10, value=4, step=1, label="Beam Size")
             submit_button = gr.Button("Submit")
         with gr.Column():
             output_text = gr.Textbox(label="Output")
 
-    submit_button.click(fn=inference, inputs=[prompt_input, image_input], outputs=output_text)
+    submit_button.click(
+        fn=inference, 
+        inputs=[prompt_input, image_input, temperature_input, beam_size_input], 
+        outputs=output_text
+    )
 
 demo.launch(share=True)